Comments from ordinary users on social media platforms can be a force for good, helping others identify false information and navigate what is accurate, new research shows.
However these same comments can also mislead when they are inaccurate, which makes it challenging for people to judge what can be trusted.
The study shows digital media literacy is not only about telling true from false but also about evaluating the reliability of user comments.
In a new book, The Power of the Crowd, Professor Florian Stöckel, from the University of Exeter, and his co-authors uncover this double-edged pattern. The book also shows how difficult it is for people to separate accurate from inaccurate information.
As part of a large-scale study with more than 10,000 participants across Germany, the UK, and Italy, researchers examined people's ability to classify true and false news in social media posts. Participants were shown a set of false and true news posts drawn from real online content.
The study included forty-seven different topics, including health, technology, and politics, all drawn from real online content. False news posts came from material flagged by fact-checking organizations in each country. The findings highlight just how challenging it is for people to identify false information: most false news stories were considered accurate by at least three out of ten people, and some were judged true by around half of respondents.
The study reveals that comments from everyday users can act as signals for others, pointing to what is true and what is not. When accurate, these comments help users spot misinformation. But when misleading, they can undermine trust in correct information.
Professor Stöckel said: "We found that user comments function like quick warning signals. People process them in a rather superficial way instead of engaging in deeper reasoning. That makes them useful when they are right, but also explains why inaccurate comments mislead so easily."
Encouragingly, the research also shows that the public broadly supports correcting false content. Survey data from Germany, for example, show that 73% of respondents prefer content to be corrected even if doing so draws more attention to the original misinformation. This is promising for those who consider whether to get involved in online debates: others are likely to appreciate it when false content is flagged.
The book also offers practical advice on how to write effective corrections. People do not need to compose lengthy comments; even short statements can be effective. What matters most is that those who post corrections get their facts right. Before posting a correction, it may be worth double-checking, for example by consulting the website of a fact-checking organisation.
Professor Stöckel said: "The potential of corrective comments lies in the fact that they offer all users a way to improve the information environment on social media even if platforms do not act."
The research also shows that people are more likely to believe false news when it aligns with their prior attitudes. The authors accounted for this in their analyses and still found small but consistent effects of corrective comments across countries.
The fieldwork was carried out in 2022 and 2023. It included posts on public health (COVID-19, vaccines, smoking), technology (the 5G cellphone network), climate change, and politics. Around 1,900 people in Britain, 2,400 in Italy, and 2,200 in Germany took part in the initial study, with an additional 4,000 people in Germany participating in a follow-up survey.
